Health Salon

Your Source for Cutting Edge Information in Alternative Health Care thats hard to find.




  • Subscribe

    • AddThis Feed Button

Links

Climate change and dietary choices —

4th March 2007 by Arrow Durfee Posted in Uncategorized

This article explains a lot about how greenhouse gases are produced from our meat industries around the world. I posted the article here because such good articles have a way of disappearing after a time. I suggest you cut and paste and to read it from where I got it. It will include graphs that I could not bring over… sorry. I’m not so good with computers.
Arrow..

**************************

seminar.mannlib.cornell.edu/topics/food_security/resources/FS01.pdf

Food Policy, Vol. 23, No. 3/4, pp. 277–293, 1998
 1998 Elsevier Science Ltd. All rights reservedPergamon
Printed in Great Britain

PII: S0306-9192(98)00037-2
Climate change and dietary choices —
how can emissions of greenhouse gases
from food consumption be reduced?
Annika Carlsson-Kanyama*
Natural Resources Management, Department of Systems Ecology, Stockholm University,
106 91 Stockholm, Sweden
Results from an analysis of greenhouse gas emissions and energy consumption during the life-cycle of carrots, tomatoes, potatoes, pork, rice and dry peas consumed in Sweden are presented and discussed. The life-cycle is delineated to the part of the production chain prior to purchase by the consumer. The study shows that emissions,
expressed in g CO2 equivalents, are highest for pork and rice and lowest for potatoes, carrots and dry peas. The most important stages of emissions in the life-cycle are identified for each of the different food items. Crop farming is the most important stage for rice and tomatoes while rearing of animals is the most important stage for pork and storage is the most important stage for carrots. Comparison with an energy analysis shows that important stages in the life-cycle of food may be under-evaluated when energy only is accounted for. This may lead to a sub-optimisation of pollution
control exemplified by the case of transportation. Also, it is shown that the choice of functional unit has a decisive influence on the outcome of the study. The recommen- dation is to compare whole meals, or diets with the same nutritional qualities. A com- parison of four meals composed of the food items under analysis shows that a meal
with tomatoes, rice and pork has nine times higher emissions than a meal made from potatoes, carrots and dry peas. Emissions of greenhouse gases from consumption patterns based on the food items analysed are compared with an assumed sustainable
limit of greenhouse gas emissions. The conclusion is that current food consumption patterns in the developed countries exceed the level of sustainability by at least a factor of 4. Prospects for achieving sustainable food consumption patterns are questionable in
view of current trends in food demand. 

1998 Elsevier Science Ltd. All rights reserved.
Keywords: food consumption, climate change, life-cycle, food, greenhouse gases, sustainability

Introduction
Climate change has emerged as one the most controversial issues on the global environmental agenda. The recent Conference of the Parties to the Framework on Climate Change in Kyoto in December 1997 managed only, after threats of breakdown, to reach an agreement whereby
the developed countries shall reduce their aggregate anthropogenic carbon dioxide equivalent

*Tel.: 46 816 42 52; Fax: 46 8 15 84 17; e-mail: annikac@system.ecology.su.se
277

278 Climate change and dietary choices: A. Carlsson-Kanyama
emissions slightly during the next decade (United Nations, 1997)1. This agreement can be viewed with the perspective that the Intergovernmental Panel on Climate Change (IPCC) con- siders that accumulated emissions of greenhouse gases have already caused a global mean surface air-temperature increase of between 0.3 and 0.6°C during the past century, and ‘that
the balance of evidence suggests a discernible human influence on the global climate’ (IPCC, 1996b, p. 4). The climate is expected to change with an average rate of warming greater than any seen in the last 10,000 years. The average sea level is expected to rise, some places are likely to experience more severe droughts and/or floods, and a sustained climate change could
lead to a shift in the competitive balance among species. For any stabilisation of the levels of CO2 in the atmosphere to occur, global anthropogenic carbon dioxide (CO2) emissions will have to drop sharply as compared to the levels of 1990. A stabilisation of atmospheric concen- trations of methane (CH4) and nitrous oxide (N2O) at today’s levels would require reductions
in anthropogenic emissions of 8% and more than 50% (IPCC, 1996a).

Emissions of the major greenhouse gases CO2, CH4, N2O and hydrofluorcarbons (HCFCs)
are closely associated with food production and consumption. During the life-cycle of food numerous human-induced activities cause emissions of these gases. For example, the use of fossil fuel in agriculture causes net emissions of CO2 as the cultivation of organic soils often does. The fabrication and application of N-fertilisers causes emissions of N2O. From the diges-
tive tracts of cows, sheep and pigs, CH4 is released. Refrigerants used in cold storage facilities often have a Global Warming Potential (GWP) and when leaked they contribute to climate change. Thus, food consumption directly and indirectly causes emissions of both energy related greenhouse gases and greenhouse gas emissions from other sources.

The consumption of food has been identified as one of the most resource demanding or polluting activities within a household (Vringer and Blok, 1995, p. 895, Table 2; National Consumer Agency, 1996, p. 11, Fig. 1; Wackernagel and Rees, 1996, p. 83).

As food is a basic need, the dietary choices of individuals and households certainly are, and will continue to be, under close scrutiny during the search for sustainable consumption patterns. Food is not easily dematerialised and cannot be substituted for services, commonly proposed as ways to lessen the environmental impacts from consumption of other products than food (OECD, 1996).

A change in diet is therefore one of the most important proposals for obtaining sustainable lifestyles in the developed countries.
Food consumption patterns are not only a reflection of nutritional needs, but also of prefer- ences for taste, odour and texture as well as culture and ethics. Therefore, sustainable food consumption patterns not only have to be considered in regards to pollution and waste gener-
ation, but also for all their immaterial qualities as well as their cultural acceptance. Historically, patterns of food consumption have changed from a diet based mainly on locally produced foods to a diet where exotic fruits, vegetables and spices are commonplace. What was considered as an attractive diet some decades ago may today be considered as strange or even unpalatable by many young people. Future food consumption patterns will continue to be a reflection of overall lifestyles, income levels and values, but should also reflect a growing consideration for the state of the environment.

Although political leaders have shown a lack of affirmative action on curbing the rate of 1 The agreement was that the developed countries shall reduce their aggregate anthropogenic carbon dioxide equivalent emissions by no more than 5% below 1990 levels in the commitment period of 2008–2012. For the developing countries, and countries with transitional economies, no agreements on reduction targets were reached (United Nations, 1997).

279Climate change and dietary choices: A. Carlsson-Kanyama
climate change, local and individual attentions to more environmentally friendly lifestyles and
consumption patterns are certainly increasing and encompassing initiatives that will, implicitly
or explicitly, lessen emissions of greenhouse gases as a result of changed consumption choices.
Examples of such initiatives are the household Eco-Teams within the framework of the Global
Action Plan (GAP) and the Perspective Project in the Netherlands (GAP International, 1996;
Perspective Secretariat, 1997) where households are encouraged to change their consumption
patterns in order to reduce the consumption of energy and water and their production of waste.
Although not large in scale at present, such programs could stimulate similar efforts on a larger
scale. Moral support of such efforts can be found in several international declarations, as well
as in Agenda 21. In Chapter 4 of Agenda 21, unsustainable consumption patterns in the
developed world are targeted as an area for major concern and consumer information is given
high priority (United Nations, 1992).

In this paper, I present a study with relevance to the issues mentioned above. Greenhouse gas emissions and energy consumption during the life-cycle of carrots, tomatoes, potatoes, pork, rice and dry peas consumed in Sweden have been calculated and the potential for changing food consumption patterns towards emission levels defined as ecologically sustainable are explored. The following questions will be addressed:

What stages, or processes, during the life-cycle of food contribute significantly to the total emissions of greenhouse gases?

How can a different functional unit2 of food alter conclusions about environmentally sound food choices?

What are the characteristics of food products which have low emissions of greenhouse gases during their life-cycle?

To what extent is the level of emissions of greenhouse gases from present food consumption compatible with sustainable levels of emissions and what are the prospects for the needed adjustments?

How does a greenhouse gas analysis of food compare with an energy analysis?

Materials and methods
Data
A detailed presentation of the data, methods and results used for the analysis of greenhouse gas emissions and energy consumption in the life-cycle of food is presented in Carlsson-Kany-
ama (1997, 1998). The results from the analysis of carrots and tomatoes were discussed in
detail in Carlsson-Kanyama (in press).
System boundary
A life-cycle approach was used in all the studies. The analysis was delineated to the part of
the production chain prior to the consumer’s purchase of food. System boundaries were selected
to facilitate comparisons between different food items. The system boundary for the vegetable
2
The functional unit is the unit to which all results are related. Examples of different functional units for food may
be a unit of weight or a unit of nutrients.
280 Climate change and dietary choices: A. Carlsson-Kanyama
products (carrots, tomatoes, potatoes, rice and dry peas) is shown in Fig. 1 and the same
boundary for the animal product in the study (pork) is shown in Fig. 2.
The calculations have included the following energy consumption and emissions of green-
houses gases:

All energy consumption during the part of the production chain which is covered in the
study (see Figs 1 and 2) was included. Pre-combustion energy consumption is included in
Figure 1 System boundary for the analysis of greenhouse gas emissions during the life-cycle of tom-
atoes, carrots, potatoes, rice and dry peas (thick line)
281Climate change and dietary choices: A. Carlsson-Kanyama
Figure 2 System boundary for the analysis of greenhouse gas emissions during the life-cycle of pork
(thick line). Although not indicated in this figure, emissions occuring during transportation within the
system boundary are included in calculations
the calculations except for electricity where figures include transmission losses but exclude
losses during conversion.
CO2 fuel-cycle emissions from fuels used to run lorries, planes, ships, tractors, drying plants
and fodder factories and generate the electricity used for storage, stables and electrical irri-
282 Climate change and dietary choices: A. Carlsson-Kanyama
gation pumps are included. CO2 emissions from the use of energy during the manufacturing
and transportation of fertilisers are also included.
Refrigerants leaking from transport refrigeration equipment in lorries, planes and ships and
from refrigeration equipment in stationary storage facilities (storage rooms) are included.
Production emissions of refrigerants are included as well.
CH4 emissions from rice fields and from the digestive tract of the pigs and their manure
are included.
Process emissions of N2O during the manufacture of fertilisers and NO2 emissions from
farm land as a result of application of N-fertilisers were included. N2O emissions during
storage of manure were included as well.
All calculated emissions of the greenhouse gases mentioned above are re-calculated into g
CO2 equivalents using the Global Warming Potentials proposed by the IPCC (1996a, p. 22,
Table 4).
Functional unit
The functional unit used in this study is one kilo of carrots, tomatoes, potatoes, rice, dry peas
or pork sold by Swedish retailers during the early or mid-1990s. These foods items had many
different countries of consumption origin (i.e. countries where the products were produced).
The results presented here have been calculated as a weighted average of emissions per kg of
food for all countries of consumption origin.
Results
Emissions of greenhouse gases and energy consumption differ greatly between the different
food items analysed. Pork emitted nine times more greenhouse gases compared to dry peas
and rice emitted 38 times more greenhouse gases than potatoes. The energy consumption for
tomatoes was almost 15 times higher than for carrots. The most important stages in the life-
cycle were, generally, crop farming and animal husbandry.
Total emissions
Table 1 shows the total greenhouse gas emissions and energy consumption for tomatoes, car-
rots, potatoes, rice, pork and dry peas. Both emissions of greenhouse gases and energy con-
sumption were lowest for the vegetable products (carrots, potatoes and dry peas). These foods
were grown in open fields and mainly in Sweden. Emissions of greenhouse gases were highest
for rice and pork; the energy consumption was highest for tomatoes followed by pork and
Table 1 Emissions of greenhouse gases in CO2 equivalents with a 20 year time perspective and energy require-
ments in MJ during the life-cycle of carrots, tomatoes, potatoes, pork, rice and dry peas consumed in Sweden.
In g CO2 equivalents per kg and MJ per kg. All emissions were supposed to occur at the same instant
Carrots Tomatoes Potatoes Pork Rice Dry peas
g CO2 500 3300 170 6100 6400 680
equivalents per
kg
MJ per kg 2.9 42 1.8 32 9.8 3.2
283Climate change and dietary choices: A. Carlsson-Kanyama
rice. Most tomatoes were assumed to be grown in greenhouses in Sweden or nearby countries,
while rice was produced in countries distant from Sweden on irrigated fields.
Relative contributions of greenhouse gases during the different stages in the life-cycle of
vegetable products
The relative contributions of the different stages in the life-cycle to the total emissions of
greenhouse gases in CO2 equivalents for the vegetable products analysed are shown in Fig. 3.
Emissions during crop farming were dominant for tomatoes, rice and dry peas, contributing
to more than 80% of the total. The emissions were caused by different types of greenhouse
gases, however. While almost all emissions during crop farming of tomatoes were caused by
the combustion of fossil fuels in greenhouses, almost all emissions during crop farming of rice
were caused by CH4 from irrigated rice fields. It is important, however, to emphasise that the
calculated emission levels of CH4 from rice farming have substantial uncertainties associated.
During crop farming of dry peas most of the emissions were caused by N2O formation resulting
from N-fixation by the crops themselves.
Emissions during storage contributed to over 60% of the total emissions for carrots while
for the other foods analysed storage caused less than 10% of the total. The reasons for the
large relative contribution of storage during the life-cycle of carrots are that carrots are com-
monly stored for a long time (several months) and also that they are cold-stored in facilities
where leakage of refrigerants are known to occur. A different situation is found for potatoes,
a food item stored over the whole winter in Sweden. Potatoes are cooled with air only during
Figure 3 Distribution of greenhouse gas emissions during the life-cycle of carrots, tomatoes, potatoes,
rice and dry peas. In CO2 equivalents with a 20 year time perspective. All emissions were supposed to
occur at the same instant
284 Climate change and dietary choices: A. Carlsson-Kanyama
storage and therefore the relative contribution of storage emissions is much lower than for car-
rots.
Emissions during transportation were 16 and 31% of the total for carrots and potatoes,
respectively, while for rice, dry peas and tomatoes transportation amounted to less than 10%
of the total. Most of the carrots, all potatoes and all the dry peas were grown in Sweden, while
all rice and 75% of the tomatoes were imported. Rice was mainly transported by ship while
the main mode of transportation for tomatoes was lorry. All Swedish grown products were
transported by lorry. Distances from the Swedish producer to the retailer were calculated to
be 650 km for carrots and 370 km for potatoes, while the distance from the rice fields in the
United States to the Swedish retailer was calculated to be 12,000 km. In absolute terms, emis-
sions per kg of crop for transportation were 420 g of CO2 equivalents for rice from the United
States and 54 g of CO2 equivalents per kg of potatoes. The reason for the large relative contri-
bution of transportation in the life-cycle of potatoes as compared to rice is, of course, that
other stages during the life-cycle were so different in magnitude.
Emissions during production and transportation of fertilisers contributed to less than 5% for
carrots, rice, tomatoes and dry peas while it was 21% for potatoes. These emissions were
mainly caused by the production and transportation of N-fertilisers.
Relative contributions of greenhouse gases during the different stages in the life-cycle of
pork
The relative contributions from the different stages in the life-cycle of pork are shown in Fig.
4. The stages with the largest were rearing of the pig and crop farming of pig feed with 34
and 24% of the total emissions, respectively.
Emissions during crop farming were lowest for the cereals (wheat, barley and oat) with
about 170 g of CO2 equivalents per kg of crop harvested. For the nitrogen fixing crops in the
study (dry peas and soya beans) emissions were 400–500 g of CO2 equivalents per kg of crop
harvested. The main reason for the higher emissions per kg of crops for the nitrogen fixating
crops was the N2O emissions related to fixation of nitrogen. However, the nitrogen fixed by
such crops is beneficial later in the crop rotation, an effect not taken into account in this study.
Emissions during rearing of the pigs were mainly caused by CH4 and N2O emissions from
the storage of pig manure, which is supposed to be stored in solid form. Seventy percent of
the emissions during pig rearing can be attributed to manure storage. Only 3% of the emissions
during pig rearing are caused by the use of electricity and fuel oil in the stables.
Emissions during production and transportation of the fertilisers (used for farming of the
crops required for production of pig fodder) were 18% of the total, with N-fertiliser related
emissions being the main contributor. Emissions during processing of fodder crops (drying of
crops, oil extraction and fodder production) were 7% of the total. Emissions during processing
were caused by energy utilisation only. Emissions during slaughtering were 2% of the total,
while transportation contributed 5%. Storage of pork contributed 10%, with storage at the
retailer as the major contributor.
Relative contributions of the different stages in the life-cycle of rice and pork to the total
energy consumption
The relative contributions of the different stages in the life-cycle when it comes to energy
consumption are shown for rice and pork only (Fig. 5).
Fig. 5 shows that the most energy demanding stages in the life-cycle of pork are crop
farming and production of fertilisers. Rearing of pigs, the most dominant stage in the life-
285Climate change and dietary choices: A. Carlsson-Kanyama
Figure 4 Distribution of greenhouse gas emissions during the life-cycle of pork. In CO2 equivalents
with a 20 year time perspective. All emissions were supposed to occur at the same instant
cycle of pork when the total greenhouse gas emissions were accounted for, was much less
significant when energy consumption was the parameter for comparison. Thus, in the energy
analysis, rearing of pigs was only 13% of the total while in the greenhouse gas analysis it was
34%. The results from analyses based on the two different environmental parameters were also
different for fodder production, drying and oil extraction. These stages were only 7% of the
total in the analysis of greenhouse gas emissions, but accounted for 17% of the total energy
consumption. The explanation of these differences is that emissions from non-energy related
greenhouse gases, such as CH4 and N2O, are important in the life-cycle of pork.
The most energy demanding stages in the life-cycle of rice are transportation (47%) while
crop farming was much less significant (19%). This can be compared to the analysis of green-
house gases for rice where transportation only accounted for 6% and crop farming 87%. Again,
the explanation for these differences is that the greenhouse gas analysis includes emissions
from many more sources than only the energy related ones, in this example mainly CH4 from
irrigated rice fields.
Interpretation of results
The results presented above may be considered as a first attempt to perform a fairly complete
estimate of emissions of greenhouse gases during the life-cycle of some foods. Inventory data
needs to be improved and expanded. Uncertainties are probably large. The typical uncertainties
286 Climate change and dietary choices: A. Carlsson-Kanyama
Figure 5 Distribution of energy requirements during the life-cycle of pork and rice
of the Global Warming Potentials only are of the order of 35%. The sensitivity of the results
to the assumptions made needs to be explored further.
Comparison of greenhouse gas analysis and energy analysis
One of the conclusions that can be drawn from my analysis is that transportation was not a
large contributor to the total emissions of greenhouse gases although it often accounted for a
major part of the total energy consumption. Furthermore, the analysis of greenhouse gases
showed that stages such as storage, pig rearing and rice farming were major contributors, but
this was not evident from an energy analysis.
Significant potential impacts caused by emissions of the non-energy related greenhouse gases
methane, nitrous oxides, hydrofluorcarbons and fluorcarbons are omitted in conventional energy
analysis of food. In view of the fact that these gases are responsible for close to 40% of the
direct radiative forcing (calculated from IPCC, 1996a) it seems plausible to conclude that an
environmental analysis of food should instead be based on a greenhouse gas emission account
rather than an account of energy consumption.
Up until now, however, efforts to improve the environmental performance of food during
the life-cycle have mainly been based on energy analysis. This has caused much focus on
transportation and little focus on, for example, storage, rearing of animals and cultivation
of irrigated rice3. Some examples will illustrate how the focus on transportation and energy
3
One example of a study where transportation has been the main target for environmental improvments is the annual
publication by the Swedish Society of Nature Conservation (1996). The title was Food and the Environment and in
Chapter 3, ‘Long road from earth to table’, it is written that ‘a neighbouring farmer is worth more than half a metre
287Climate change and dietary choices: A. Carlsson-Kanyama
consumption can sometimes lead to minor improvements only, while a focus on greenhouse
gases could bring about more drastic emission cuts.
If energy consumption for transportation is lowered by 30% in the life-cycle of rice, total
emissions of greenhouse gases only will decrease by less than 2%. Conventional energy analy-
sis would indicate a decrease in the total emissions connected to the energy consumption by
14% for rice. If energy consumption in the pig house is lowered by 50%, total emissions of
greenhouse gases in the life-cycle of pork will decrease by less than 1% while an energy
analysis would indicate a decrease of 6%. If emissions of methane during rice farming are
lowered by 30%, this would lead to a decrease in the total emissions of greenhouse gases,
expressed in CO2 equivalents, by 23% for rice. The latter measure may prove to be more cost-
efficient than lowering energy consumption for transportation. Hence, an analysis with a single
numerator such as energy or monetary costs may omit important information for decision-mak-
ing.
Comparison of analyses using different functional units
The greenhouse gas emissions and energy consumption per kg of food analysed are presented
in Table 1. Table 2 shows the nutritional qualities per kg of the same food items as well as
a comparison with the results from Table 1 if the functional units used in the analysis were
the quantity of energy (MJ/kg), or the contents of protein (g/kg) or vitamin β-carotene (μg/kg)
in the food instead of a unit of weight (kg of final product). As can be seen from Table 2,
conclusions about the most emitting or lowest emitting food depend entirely on the functional
unit chosen.
The most efficient food when only the energy contents are compared is dry peas followed
by potatoes and carrots. Tomatoes, with high emissions due to abundant energy requirements
during greenhouse cultivation, are highly inefficient as energy suppliers. However, when a
different basis of comparison is used — vitamin β-carotene — tomatoes are found to be
superior to all other foods analysed, except carrots which are extremely rich in vitamin β-
carotene. Since neither rice nor pork contain any vitamin β-carotene, no values of CO2 equiva-
lents emitted per μg vitamin β-carotene can be calculated. Most efficient for producing protein
Table 2 Nutrient contents of some food (Swedish National Food Administration, 1996) and the emissions of
greenhouse gases per unit of nutrient
Food MJ per kg Protein β-carotene g CO2 g CO2 g CO2
(g per kg) (μg per kg) equivalents equivalents equivalents
per MJ per g protein per μg β-
carotene
Tomatoes 0.83 9 5730 4000 370 1
Carrots 1.67 6 68,000 300 83 0
Potatoes 3.1 18 100 56 10 2
Rice 14.9 68 0 430 94 —
Pork 7.2 180 0 850 34 —
Dry peas 12.4 215 150 55 3 5
of extra insulation on the house’, p. 47. Another example of how transportation is targeted is found in a publication
from the local environmental protection agency in Stockholm (Agenda 21 Miljo ̈ fo ̈ rvaltningen i Stockholm, 1996). In
this publication, the energy requirements for transportation of different foods are calculated with the conclusion that
locally produced food is more environmentally benign.
288 Climate change and dietary choices: A. Carlsson-Kanyama
with respect to emissions of greenhouse gases are dry peas, followed by potatoes and pork.
Thus, by changing the functional unit used for comparing different food items, conclusions
about which type of food that is the least or the most environmentally polluting can be
quite different.
Characteristics of food with low emissions of greenhouse gases during their life-cycle
Thus, a preferable choice of a functional unit for food may be to use the whole diet or meals
with similar nutritional qualities, but even such a comparison is incomplete because the many
immaterial qualities of food, such as taste, colour, odour and texture still have to be taken
into account.
In Fig. 6, the ingredients of four different meals which contain the same amount of energy
and protein, but have very different potentials for changing the global climate, are shown. Fig.
7 shows the emissions of greenhouse gases in g CO2 equivalents for each of the different meals.
The meals a–d in Fig. 6 can be characterised in the following manner. Meal a is purely
vegetarian with ingredients mainly from the domestically produced foods carrots, potatoes and
dry peas. Meal b is also vegetarian, but with the exotic ingredients rice and tomatoes in addition
to dry peas. Tomatoes are considered exotic because either they have to be imported from
southern Europe or else grown in greenhouses. Meal c is solely based on animal and exotic
foods: rice, tomatoes and pork while meal d contains only domestically produced food of both
animal and vegetable origin. The meals differ with regards to amounts of ingredients also. The
ingredients of the vegetarian exotic meal (meal b) weigh only 250 g while the ingredients of
Figure 6 Ingredients of four different meals with the same energy and protein contents (2 MJ and 22–
24 g of protein). In kg ingredient per type of food and meal
289Climate change and dietary choices: A. Carlsson-Kanyama
Figure 7 Emissions of greenhouse gases in CO2 equivalents from four different meals with the same
energy and protein contents (2 MJ and 22–24 g of protein). In g CO2 equivalent with a 20 year time
perspective according to meal and type of ingredient
the vegetarian domestic meal (meal a) and the mixed domestic meal (meal d) weigh about
500 grams per meal. These differences influence the magnitude of greenhouse gas emissions
from transportation of the ingredients from the retailer to the final consumer, but it has been
assumed that these emissions are not large enough to alter the results shown in Fig. 74.
In Fig. 7 it is shown that emissions of greenhouse gases during the life-cycle of the four
meals differ substantially. The animal–exotic meal (meal c) emits 1800 g of CO2 equivalents
while the vegetarian–domestic meal (meal a) only emits 190 g of CO2 equivalents. This makes
the animal–exotic meal nine times more polluting than the vegetarian–domestic meal. This
does not mean that vegetarian food is generally less polluting than animal food. The results
in Fig. 7 show that vegetarian food consumption patterns (meal b), may, in some cases, be
much more polluting than meals with animal ingredients (meal d). The vegetarian–exotic meal
(meal b) emitted 860 g of CO2 equivalents as compared to only 380 g of CO2 equivalents for
meal d.
4
For example, assuming that a car consuming 0.9 liters of petrol per 10 km is used for shopping, that the distance to
the shop is 5 km and that 30 kg of food is bought. The subsequent emissions of CO2 may be 91 g of CO2 per kg of
food transported (0.9 liters 38 MJ 80 g of CO2 per MJ/30 kg). For the meals a and d that would mean an
additional 45 g of CO2 equivalents, respectively, and for meal b an additional 23 g of CO2 equivalents.
290 Climate change and dietary choices: A. Carlsson-Kanyama
Current emission levels from food consumption and prospects for sustainability
Sustainable consumption patterns have, for the purpose of the following discussion, been
defined according to the following criteria:
The accumulated emissions of anthropogenic CO2 from now until 2100 must not exceed a
level at which the atmospheric concentration of CO2 is expected to stabilise at 450 ppmv.
This means that accumulated anthropogenic emissions over the period from 1991 to 2100
should not exceed 460 GtC (IPCC, 1996a, Summary for Policymakers).
Emissions of CH4 and N2O should be stabilised at today’s levels (1995). This would involve
reductions in current anthropogenic emissions of these gases by at least 8% for CH4 and
more than 50% for N2O (IPCC, 1996a, p. 26). Current anthropogenic emissions of CH4
have been estimated at 375 Tg per year and the same emissions of N2O are estimated to be
3–8 Tg N per year (IPCC, 1996a, pp. 18–19).

Every person now living on the Earth and all those expected to live until 2100 should have
the same rights to emit anthropogenic CO2 and other greenhouse gases. For this purpose,
the total number of human years expected to occur from 1991 to 2100 were calculated based
on recent population estimates (United Nations, 1995, medium variant of the population
projection) and was found to be 977 G human years. Present world population was assumed
to be 5.89 G people (United Nations, 1995).

Results from calculations of a sustainable limit for consumption are shown in Table 3 where
a total of 5900 kg CO2 equivalents per capita and year represent the sustainable limit for
consumption. This ‘greenhouse gas emission space’ can be divided into many consumption
categories. As a tentative division, let us assume 30% of the greenhouse gas emission space
is set aside for common purposes, such as defence, education and health care. Even today,
any citizen in any country has to set aside a certain share of his or her monetary income for such
purposes. The remaining 70% of the greenhouse gas emissions space, 4100 kg, is supposed to
include emissions caused by all other consumption, such as buying and eating food, travelling
and heating the home. Assuming that 25% of the remaining space is allocated to food consump-
tion, the sustainable level would be about 1000 kg of CO2 equivalents per capita and year.
The daily intake of food energy in Sweden has been about 12 MJ per person and day during
the last 15 years (Swedish Board of Agriculture, 1997, p. 29). Assuming that all this energy
intake is supplied from one of the four different meals presented earlier (meals a–d), the
conclusions are as follows.

If all the energy intake during 1 year is supplied by food as in meal a or d, the emissions
of greenhouse gases would be 420 and 830 kg of CO2 equivalents, respectively, which is well
within the limits of the estimated greenhouse gas emission space. Meals a and d were both
based on vegetable products of the season complemented with pulses and small amounts of
pork.

Table 3 The greenhouse gas emissions space for CO2, CH4 and N2O. In kg greenhouse gases and kg CO2
equivalents per capita and year Permitted emissions of Permitted emissions of Permitted emissions of Permitted emissions of
CO2 per capita and year CH4 per capita and year N2O per capita and year CO2 equivalents with a
1991–2100 20 year time perspective per capita and year 2400 kg 59 kg 0.67 kg 5900 kg
291Climate change and dietary choices: A. Carlsson-Kanyama
If, on the other hand, all the energy intake is supplied by food as in meal b or c, the emissions
of greenhouse gases would be far above the limit of the greenhouse gas emission space. The
emissions from a diet of food as in meal b (vegetarian–exotic) is 1900 kg of CO2 equivalents
per capita and year. For a diet of food as in meal c (animal–exotic) the emissions amount to
3800 kg of CO2 equivalents. How large the emissions from present Swedish food consumption
patterns are in reality remains to be answered, but in view of the fact that animal products
contribute to about 30% of the energy supply (Swedish Board of Agriculture, 1997, calculated
from p. 31), it seems plausible that present Swedish consumption patterns resemble a diet that
causes emissions of the same amount per year as meal c rather than meal a or d. This means
that the greenhouse gas emissions from current food consumption patterns exceed the sus-
tainable level by a factor of four.

The prospects for achieving sustainability of such consumption patterns seem questionable.
The relationships between standard of living and diet have been reported as predictable and
strong with the importance of starchy staples declining as income grows and the importance
of ‘luxury food’ increasing (Bender, 1994). In a more prosperous world with an increasing
population this seem not to bode well for climate change.

The advocates for more sustainable consumption patterns certainly have many well-founded
reasons other than those related to environmental impacts for their claims. For example, a
move towards a more vegetable diet in the developed world can be recommended because a
substantial amount of grain could then become available for human consumption. This would
improve the prospective for supplying a growing world population with an adequate food
supply (Kendall and Pimentel, 1994). Also, increased health and obesity problems in the
developed parts of the world may be forceful agents for preventing over-consumption. Already
such concerns have contributed to changes in the mix of meat consumption towards leaner
meats (Bender, 1994). Leaner meat, such as chicken, usually has higher feed conversion
efficiencies than red meat, such as beef.

Future food consumption patterns in the developed parts of the world may develop along
several different lines. Generally, the demand for convenience food is expected to increase as
are concerns for health and the environment (Popcorn, 1991). Meat consumption, and especially
the consumption of red meat, may continue to decline because of changing perceptions of the
world (Fiddes, 1991). What the consequences in terms of emissions of greenhouse gases will
be is unclear. As was shown in this paper, consumption patterns that are purely vegetarian
may still be unsustainable. End-use efficiency of food, which is today low, (Bender, 1994)
could increase or decrease as a result of the trend for more convenience food.

Conclusions
Depending on the environmental parameter chosen in an analysis, conclusions about which
stage or process to target in the life-cycle of food for pollution control may be quite different.
Thus, transportation may be one of the main targets when a conventional energy analysis
is used, but not when greenhouse gas emissions from all sources are accounted for. When
a conventional energy analysis is used for food, important emissions which affect the climate
are not included. This may lead to a sub-optimisation of pollution control.

Conclusions about sustainable food consumption based on a single functional unit such as
weight or energy content of the food itself must be drawn carefully, otherwise recommen-
dations may be skewed. An example of such a misinterpretation is a recent article where
292 Climate change and dietary choices: A. Carlsson-Kanyama
MJ was used as the only functional unit with the result that potato chips, white bread and
ice-cream were regarded as more environmentally benign than fruit, high-fibre cereals and
meat (New Scientist, 1997, p. 10). This result is not surprising as chips contain 19 MJ per
kg (and lots of fat too) as compared to apples which only contain 2.2 MJ per kg (with very
little fat but lots of vitamins).

Food consumption patterns which are low in emissions of greenhouse gases have low shares
of animal food and low shares of what, in Sweden, may be considered as exotic ingredients.
Current Swedish consumption patterns, and most certainly consumption patterns in other
parts of the developed world as well, exceed the greenhouse gas emission space with a
magnitude of at least a factor of 4. A diet within the limits of the greenhouse gas emission
space could, for example, be composed of a small amount of animal products complemented
with vegetable protein sources and vegetables of the season. Many of the current trends in
food demand seem to be counteracting eventual improvements in technology, such as more
fuel efficient lorries, more energy efficient refrigerators and more efficient farming practices.
Overall, it seems that technological changes will have to be very profound for reducing
current greenhouse gas emissions from food consumption to a sustainable level. Whether
those radical changes in food demand that are needed for ecological reasons will occur or
not is an open question. Clearly, the future will bring tremendous challenges to every actor
connected to food manufacturing, farming or consumption as well as to researchers within
all fields.

Acknowledgements
I would like to thank my colleagues and supervisors at the Department of Systems Ecology,
Stockholm University, who made useful suggestions on how to improve the quality of this
manuscript. This work was financed by the Swedish Waste Research Council and the Foun-
dation for Strategic Environmental Research.
References
Agenda 21 Miljo ̈ fo ̈ rvaltningen i Stockholm (1996) Livsmedelstransporter och miljo ̈ Del 1 och 2. Transportation of
Food Part 1 and 2. In Swedish.
Bender, W. H. (1994) An end use analysis of global food requirements. Food Policy 19, 381–395.
Biesot, W. and Moll, H. C. (1995) Reduction of CO2 emissions by lifestyle changes. Final Report to the NRP Global
Air Pollution and Climate Change. Center for Energy and Environmental Studies, IVEM, Groningen University,
IVEM research report nr. 80, Netherlands.
Carlsson-Kanyama, A. (1997a) Food and the environment: implications of Swedish consumption patterns. Licentiate
Thesis, Department of Environmental and Energy Systems Studies, Lund University, Sweden.
Carlsson-Kanyama, A., Food consumption patterns and their influence on climate change—greenhouse gas emissions
in the life-cycle of tomatoes and carrots consumed in Sweden. Accepted in Ambio (in press).
Carlsson-Kanyama, A. (1998) Energy consumption and emissions of greenhouse gases in the life-cycle of potatoes,
pork meat, rice and yellow peas. Technical Report no. 26, Department of Systems Ecology, Stockholm Univer-
sity, Sweden.
GAP International, GAP’s EcoTeam program is a community-based campaign. Leaflet from GAP International,
Enebyberg, Sweden, 1996.
IPCC, Intergovernmental Panel on Climate Change, Climate Change 1995, The Science of Climate Change, Cambridge
University Press, 1996a.
IPCC, Intergovernmental Panel on Climate Change, Revised 1996 IPCC Guidelines for Greehouse Gas Inventories:
Reference Manual. Available through OECD, Paris, 1996b.
Fiddes, N., Meat A Natural Symbol, Routledge, New York, 1991.
Kendall, H. W. and Pimentel, D. (1994) Constraints on the expansion of the global food supply. Ambio 23, 198–205.
National Consumer Agency, Miljobelastning ved familiens aktiviteter. Resultater fra en kortlaegning af ressourcefor-
293Climate change and dietary choices: A. Carlsson-Kanyama
brug og udledninger till miljoet fra 22 aktiviteter i ne modelfamilie. (Environmental loads from family activitities.
Results from a survey of resource requiremenst and emissions to the environment from 22 activities in a model
family). Copenhagen, Rapport 1996:1. In Danish. The report could be downloaded from
www.forbrugerstyrelsen.dk/ on 970506.
New Scientist, White bread is green. Suprisingly, processed food is easier on the planet. 6 December 1997.
OECD, OECD news release from the meeting of OECD environmental policy committee at ministerial level, Paris
19–20 February 1996, SG/COM/NEWS (96) 15, Paris, 1996.
Perspectief Secratariat, Can we save energy by changing our lifetstyles? Leaflet from the Perspectief Secretariat,
Rotterdam, The Netherlands, 1997.
Popcorn, F., The Popcorn Report, Doubleday, New York, 1991.
Swedish Board of Agriculture, Livsmedelskonsumtionen 1993–1996. Food consumption 1993–1996, Rapport 1997:15,
Jo ̈ nko ̈ ping, Sweden, 1997. In Swedish.
Swedish National Food Administration, Livsmedelstabeller energi och na ̈ ringsa ̈ mnen, Tables of food energi and nutri-
ents, Jo ̈ nko ̈ ping, Sweden, 1996. In Swedish.
Swedish Society of Nature Conservation, Maten och miljo ̈ n, Råvaror-resurser-recept, Food and the Environment, Raw
Materials-Resources-Recepies, Stockholm, 1996. In Swedish.
United Nations, Agenda 21, United Nations Conference on Environment and Development, Rio de Janeiro, 3–14 June
1992, Chapter 4, a/conf. 151/4, United Nations, New York, 1992.
United Nations, World Population Prospects, The 1994 Revision, Department for Economic and Social Information
and Policy Analysis Population Division, New York, 1995.
United Nations, Kyoto Protocol to the United Nations Framework Convention on Climate Change,
FCCC/CP/1997/L.7/Add.1, 10 December 1997. Could be downloaded from
www.cop3.de/fccc/docs/cop3/107a01.pdf on January 1998.
Vringer, K. and Blok, K. (1995) The direct and indirect energy requirements of households in the Netherlands. Energy
Policy 23, 893–910.
Wackernagel, M. and Rees, W., Our Ecological Footprint. Reducing Human Impacts on the Earth, New Catalyst
Bioregional Series No. 9, New Society Publishers, Gabriola Island, B.C. Canada, 1996.

Related Posts:

Comments are closed.